WF03The viral 50, re-tested · Daily
Not re-testedAnchored Walk-Forward With Pre-OOS Purging
The rules
Re-checks the weekly RSI rule (W03) with purged, anchored windows.
- Universe
- XLB, XLE, XLF, XLI, XLK, XLP, XLU, XLV and XLY
- Bars
- daily
- Position size
- -
- Rebalance
- -
Results
| Measure | Orbital re-test | Published | SPY, same days |
|---|---|---|---|
| Window | Not re-tested | — | — |
| Total return | — | — | — |
| CAGR | — | — | — |
| Sharpe | — | — | — |
| Discounted Sharpe | — | — | — |
| Max drawdown | — | — | — |
| Win rate | — | — | — |
| Profit factor | — | — | — |
| Trades | — | — | — |
Rebuild it in Orbital
Paste this into Orbital Chat. The Strategy Developer builds the workflow and asks before it runs any backtest.
Closest Orbital version: this library's version of W03's Friday RSI rule (25% of equity per ETF, up to four, most oversold first) on three training windows that start in October 2016 and grow by two years, each followed by a two-year test. Each window is its own backtest from cash, six in all where the original's 24-month and 6-month windows would need more than 30, and the six-to-ten-week gaps are the original's purge of 10 trading days plus 2% of the training window, which grows with it.
Trade the sector ETFs XLB, XLE, XLF, XLI, XLK, XLP, XLU, XLV and XLY on daily bars, deciding on Fridays only. Buy an ETF with 25% of equity when RSI(14) closes below 35, most oversold first, and sell it when RSI(14) closes above 60. Long only; uninvested capital stays in cash. Then check it with an anchored walk-forward, keeping the rule the same in every window: three training windows of about four, six and eight years, all starting on 2016-10-31, each stopping six to ten weeks before the two-year test window that follows it. Backtest each window separately with $100,000: training 2016-10-31 to 2020-09-18 and test 2020-11-02 to 2022-10-28; training 2016-10-31 to 2022-09-01 and test 2022-10-31 to 2024-10-30; training 2016-10-31 to 2024-08-20 and test 2024-10-31 to 2026-10-30. That is six backtests. When all six have finished, compare them in one table with annual return, Sharpe and maximum drawdown, and say whether each test window kept up with its training window and whether the test Sharpe fell from one window to the next.
Why it was not re-tested
A validation workflow around W03's rule rather than a strategy of its own. Orbital covers part of the same ground for every strategy: its Backtesting list discounts the Sharpe for every configuration tried, and each report resamples the trades 1,000 times.
Sources